A note on the local behavior of the Taylor method for stiff ODEs

IF 3.4 2区 数学 Q1 MATHEMATICS, APPLIED Applied Mathematics and Computation Pub Date : 2025-02-13 DOI:10.1016/j.amc.2025.129344
Philip P. Forrier , Joan Gimeno , Àngel Jorba
{"title":"A note on the local behavior of the Taylor method for stiff ODEs","authors":"Philip P. Forrier ,&nbsp;Joan Gimeno ,&nbsp;Àngel Jorba","doi":"10.1016/j.amc.2025.129344","DOIUrl":null,"url":null,"abstract":"<div><div>In this note we study the behavior of the coefficients of the Taylor method when computing the numerical solution of stiff Ordinary Differential Equations. First, we derive an asymptotic formula for the growth of the stability region w.r.t. the order of the Taylor method. Then, we analyze the behavior of the Taylor coefficients of the solution when the equation is stiff. Using jet transport, we show that the coefficients computed with a floating point arithmetic of arbitrary precision have an absolute error that depends on the variational equations of the solution, which can have a dominant exponential growth in stiff problems. This is naturally related to the characterization of stiffness presented by Söderlind et al. <span><span>[32]</span></span>, and allows to discuss why explicit solvers need a stepsize reduction when dealing with stiff systems. We explore how high-order methods can alleviate this restriction when high precision computations are required. We provide numerical experiments with classical stiff problems and perform extended precision computations to demonstrate this behavior.</div></div>","PeriodicalId":55496,"journal":{"name":"Applied Mathematics and Computation","volume":"496 ","pages":"Article 129344"},"PeriodicalIF":3.4000,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Mathematics and Computation","FirstCategoryId":"100","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0096300325000712","RegionNum":2,"RegionCategory":"数学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MATHEMATICS, APPLIED","Score":null,"Total":0}
引用次数: 0

Abstract

In this note we study the behavior of the coefficients of the Taylor method when computing the numerical solution of stiff Ordinary Differential Equations. First, we derive an asymptotic formula for the growth of the stability region w.r.t. the order of the Taylor method. Then, we analyze the behavior of the Taylor coefficients of the solution when the equation is stiff. Using jet transport, we show that the coefficients computed with a floating point arithmetic of arbitrary precision have an absolute error that depends on the variational equations of the solution, which can have a dominant exponential growth in stiff problems. This is naturally related to the characterization of stiffness presented by Söderlind et al. [32], and allows to discuss why explicit solvers need a stepsize reduction when dealing with stiff systems. We explore how high-order methods can alleviate this restriction when high precision computations are required. We provide numerical experiments with classical stiff problems and perform extended precision computations to demonstrate this behavior.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
关于刚性ode的Taylor方法的局部行为的注释
在本文中,我们研究了在计算刚性常微分方程数值解时泰勒方法系数的行为。首先,利用泰勒方法的阶,导出了稳定区域增长的渐近公式。然后,我们分析了当方程为刚性时解的泰勒系数的行为。利用射流输运,我们证明了用任意精度的浮点算法计算的系数具有绝对误差,这取决于解的变分方程,在刚性问题中可能具有主导指数增长。这自然与Söderlind等人提出的刚度特征有关,并允许讨论为什么显式求解器在处理刚性系统时需要减少步长。我们探讨了当需要高精度计算时,高阶方法如何减轻这种限制。我们提供了经典刚性问题的数值实验,并进行了扩展的精确计算来证明这种行为。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
CiteScore
7.90
自引率
10.00%
发文量
755
审稿时长
36 days
期刊介绍: Applied Mathematics and Computation addresses work at the interface between applied mathematics, numerical computation, and applications of systems – oriented ideas to the physical, biological, social, and behavioral sciences, and emphasizes papers of a computational nature focusing on new algorithms, their analysis and numerical results. In addition to presenting research papers, Applied Mathematics and Computation publishes review articles and single–topics issues.
期刊最新文献
Nonmonotone quasi-Newton method with diagonal Jacobian approximation for symmetric nonlinear equations Event-triggered stabilization of fractional linear systems with time-varying delays: A Halanay inequality approach Observer-based event-triggered composite anti-disturbance finite-time bounded consensus control for multi-Agent systems under switching topologies and multiple attacks Editorial Board Oscillatory performance of the solutions of neutral differential equations with a delayed damping term
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1